13

I have a .dat file that contains two columns of numbers so it looks something like this:

111    112
110.9  109
103    103

and so on.

I want to plot the two columns against one another. I have never dealt with a .dat file before so I am not sure where to start.

So far I figured out that numpy has something I can use to call.

data = numpy.loadtxt('data.DAT')

but I'm not sure where to go from here. Any ideas?

martineau
  • 119,623
  • 25
  • 170
  • 301
Dax Feliz
  • 12,220
  • 8
  • 30
  • 33
  • 1
    so then you have a 2d array of points ... this has nothing to do with *.dat file could be anything *.txt would work exactly the same... your real question is "How Do I plot a numpy array?" – Joran Beasley Sep 07 '12 at 04:35
  • 1
    It's easy in `gnuplot` ;^). `plot 'yourfile.dat' u 1:2` (but of course, that doesn't address the actual question ...) – mgilson Sep 07 '12 at 04:35
  • You can use Scavis that interfaces NumPy (or JNumeric in Java) as [explained in the Scavis manual](http://jwork.org/scavis/wikidoc/doku.php?id=man:numeric:jnum) –  Jan 29 '14 at 03:36
  • Possible duplicate of [Matplotlib basic plotting from text file](http://stackoverflow.com/questions/11248812/matplotlib-basic-plotting-from-text-file) – Ciro Santilli OurBigBook.com Aug 29 '16 at 18:55

2 Answers2

15

Numpy doesn't support plotting by itself. You usually would use matplotlib for plotting numpy arrays.

If you just want to "look into the file", I think the easiest way would be to use plotfile. 

import matplotlib.pyplot as plt 

plt.plotfile('data.dat', delimiter=' ', cols=(0, 1), 
             names=('col1', 'col2'), marker='o')
plt.show()

You can use this function almost like gnuplot from within ipython:

$ ipython --pylab
...
...
In [1]: plt.plotfile('data.dat', delimiter=' ', cols=(0, 1), 
...                  names=('col1', 'col2'), marker='o')

or put it in a shell script and pass the arguments to it to use it directly from your shell

plotfile_example

bmu
  • 35,119
  • 13
  • 91
  • 108
4
import numpy as np
import matplotlib.pyplot as plot
#data = np.loadtxt("plot_me.dat")
#x,y=np.loadtxt("plot_me.dat",unpack=True) #thanks warren!
#x,y =  zip(*data)
#plot.plot(x, y, linewidth=2.0)
plot.plot(*np.loadtxt("plot_me.dat",unpack=True), linewidth=2.0)
plot.show()

[Edit]Thanks for the tip i think its as compact as possible now :P

plot 1

If you want it to be log10 just call log10 on the nparray)

plot.plot(*np.log10(np.loadtxt("plot_me.dat",unpack=True)), linewidth=2.0)

log10

Joran Beasley
  • 110,522
  • 12
  • 160
  • 179